five

open-llm-leaderboard/details_huggingface__llama-65b

收藏
Hugging Face2023-10-17 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/open-llm-leaderboard/details_huggingface__llama-65b
下载链接
链接失效反馈
官方服务:
资源简介:
该数据集是在模型 huggingface/llama-65b 在 Open LLM Leaderboard 上的评估运行期间自动创建的。数据集由 64 个配置组成,每个配置对应一个被评估的任务。数据集是从 2 次运行中生成的,每次运行在每个配置中表示为特定的分割,分割名称使用运行的时间戳。train 分割始终指向最新的结果。一个名为 results 的额外配置存储了所有运行的聚合结果,这些结果用于计算和显示 Open LLM Leaderboard 上的聚合指标。README 还提供了一个示例,展示了如何使用 `load_dataset` 函数加载运行中的详细信息。

This dataset was automatically generated during the evaluation runs of the huggingface/llama-65b model on the Open LLM Leaderboard. It comprises 64 configurations, each corresponding to a single evaluated task. The dataset is created from two runs, where each run is represented as a dedicated split under each configuration, with split names using the timestamp of the corresponding run. The 'train' split always points to the most up-to-date results. An additional configuration named 'results' stores the aggregated results across all runs, which are utilized to compute and display the aggregate metrics on the Open LLM Leaderboard. The README also includes an example showing how to use the `load_dataset` function to load the detailed information from the runs.
提供机构:
open-llm-leaderboard
原始信息汇总

数据集概述

数据集简介

该数据集是在对模型 huggingface/llama-65b 进行评估运行期间自动创建的,用于 Open LLM Leaderboard

数据集结构

  • 配置数量:64个配置,每个配置对应一个评估任务。
  • 运行次数:数据集从2次运行中创建。每个运行在每个配置中作为一个特定的分割存在,分割名称使用运行的时间戳。
  • 分割:每个配置包含多个分割,其中 "train" 分割始终指向最新的结果。
  • 结果汇总:一个额外的配置 "results" 存储所有运行的汇总结果,用于计算和显示在 Open LLM Leaderboard 上的聚合指标。

数据加载示例

python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_huggingface__llama-65b", "harness_winogrande_5", split="train")

最新结果

以下是 2023-10-17T19:28:15.120039 运行 的最新结果: python { "all": { "em": 0.0014681208053691276, "em_stderr": 0.00039210421902984954, "f1": 0.05979341442953042, "f1_stderr": 0.0013171267692059473, "acc": 0.551148504673065, "acc_stderr": 0.011494117650229113 }, "harness|drop|3": { "em": 0.0014681208053691276, "em_stderr": 0.00039210421902984954, "f1": 0.05979341442953042, "f1_stderr": 0.0013171267692059473 }, "harness|gsm8k|5": { "acc": 0.2767247915087187, "acc_stderr": 0.012323047397959794 }, "harness|winogrande|5": { "acc": 0.8255722178374112, "acc_stderr": 0.010665187902498433 } }

配置详情

以下是部分配置的详细信息:

配置:harness_arc_challenge_25

  • 分割
    • 2023_07_21T14_43_31.065297
    • latest
  • 路径
    • **/details_harness|arc:challenge|25_2023-07-21T14:43:31.065297.parquet

配置:harness_drop_3

  • 分割
    • 2023_10_17T19_28_15.120039
    • latest
  • 路径
    • **/details_harness|drop|3_2023-10-17T19-28-15.120039.parquet

配置:harness_gsm8k_5

  • 分割
    • 2023_10_17T19_28_15.120039
    • latest
  • 路径
    • **/details_harness|gsm8k|5_2023-10-17T19-28-15.120039.parquet

配置:harness_hellaswag_10

  • 分割
    • 2023_07_21T14_43_31.065297
    • latest
  • 路径
    • **/details_harness|hellaswag|10_2023-07-21T14:43:31.065297.parquet

配置:harness_hendrycksTest_5

  • 分割
    • 2023_07_21T14_43_31.065297
    • latest
  • 路径
    • **/details_harness|hendrycksTest-abstract_algebra|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-anatomy|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-astronomy|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-business_ethics|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-clinical_knowledge|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-college_biology|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-college_chemistry|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-college_computer_science|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-college_mathematics|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-college_medicine|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-college_physics|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-computer_security|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-conceptual_physics|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-econometrics|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-electrical_engineering|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-elementary_mathematics|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-formal_logic|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-global_facts|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-high_school_biology|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-high_school_chemistry|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-high_school_computer_science|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-high_school_european_history|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-high_school_geography|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-high_school_government_and_politics|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-high_school_macroeconomics|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-high_school_mathematics|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-high_school_microeconomics|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-high_school_physics|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-high_school_psychology|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-high_school_statistics|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-high_school_us_history|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-high_school_world_history|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-human_aging|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-human_sexuality|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-international_law|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-jurisprudence|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-logical_fallacies|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-machine_learning|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-management|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-marketing|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-medical_genetics|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-miscellaneous|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-moral_disputes|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-moral_scenarios|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-nutrition|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-philosophy|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-prehistory|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-professional_accounting|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-professional_law|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-professional_medicine|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-professional_psychology|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|hendrycksTest-public_relations|5_2023-07-21T14:43:31.065297.parquet
    • **/details_harness|
5,000+
优质数据集
54 个
任务类型
进入经典数据集
二维码
社区交流群

面向社区/商业的数据集话题

二维码
科研交流群

面向高校/科研机构的开源数据集话题

数据驱动未来

携手共赢发展

商业合作